I'd like to share the journey of my open-source CLI project, intelli-shell. It didn't start as a big idea, but as a simple utility to solve my own frustration.
What started as a personal tool has evolved significantly over time, and I wanted to share its progress:
*Phase 1: Building a solid foundation (v1.0)*
Initially, it was just a command snippet manager tialored to my own needs. After positive feedback, I did a full rewrite to build a solid foundation. The focus was purely on a great user experience, a clean CLI interface, and making it highly configurable. It did one thing well: save and retrieve your personal command templates.
*Phase 2: Tackling the unknown with AI (v2.0)*
Next, I tackled the problem of commands I didn't already know. In v2.0, I integrated an AI assistant. This allowed `intelli-shell` to not only retrieve saved commands but also generate, explain, or even fix new ones from a plain English prompt. It became a tool for both remembering and discovering.
*Phase 3: Bringing it to life with dynamic completions (v3.0)*
The most recent major step was adding what I think is its most powerful feature: dynamic, context-aware completions. When a command template needs an argument like a Docker container ID, a Kubernetes pod name, or a git branch, `intelli-shell` can now execute a shell command on the fly to fetch a live list of options for you to choose from. This bridges the gap between static templates and your live environment.
I'm sharing it today because I feel it has reached a level of maturity where it can be genuinely useful to other developers. I'd love to get your thoughts, critiques, and ideas.
lasantosr•1h ago
What started as a personal tool has evolved significantly over time, and I wanted to share its progress:
*Phase 1: Building a solid foundation (v1.0)* Initially, it was just a command snippet manager tialored to my own needs. After positive feedback, I did a full rewrite to build a solid foundation. The focus was purely on a great user experience, a clean CLI interface, and making it highly configurable. It did one thing well: save and retrieve your personal command templates.
*Phase 2: Tackling the unknown with AI (v2.0)* Next, I tackled the problem of commands I didn't already know. In v2.0, I integrated an AI assistant. This allowed `intelli-shell` to not only retrieve saved commands but also generate, explain, or even fix new ones from a plain English prompt. It became a tool for both remembering and discovering.
*Phase 3: Bringing it to life with dynamic completions (v3.0)* The most recent major step was adding what I think is its most powerful feature: dynamic, context-aware completions. When a command template needs an argument like a Docker container ID, a Kubernetes pod name, or a git branch, `intelli-shell` can now execute a shell command on the fly to fetch a live list of options for you to choose from. This bridges the gap between static templates and your live environment.
I'm sharing it today because I feel it has reached a level of maturity where it can be genuinely useful to other developers. I'd love to get your thoughts, critiques, and ideas.
The project is on GitHub: https://github.com/your-username/intelli-shell
Thanks for taking a look!